Eating disorders are significant mental health problems that disproportionally affect girls. The substantial psychiatric and medical morbidity resulting from these disorders attest to their public health significance and the need to understand their etiology. Research suggests that prenatal testosterone may masculinize (i.e., lower) risk for disordered eating (DE) and account for sex differences in prevalence, yet how or when these effects emerge remains unknown. Elevated prenatal testosterone decreases sensitivity to ovarian hormones during and after puberty in animals. Importantly, ovarian hormones are associated with several types of DE after puberty, thus, one potential mechanism for testosterone's effects on DE may be to decrease sensitivity to ovarian hormones during puberty. Opposite-sex twin pairs provide a novel human design for examining this hypothesis, as opposite-sex female twins develop in utero with a male co-twin and are believed to be exposed to higher levels of testosterone prenatally. This twin design will be used to: 1) examine whether the masculinizing effects of prenatal testosterone on DE emerge during puberty;2) confirm that these findings are not due to the confounding effects of other factors that change during puberty (e.g., adiposity, mood, autonomy difficulties) or to being reared with a brother. Participants (ages 10-15) will include an archival sample of same-sex male (N= 270) and female (N = 350) twins and new samples of: 1) opposite-sex twins (N = 120);and 2) non-twin females reared with brothers (N = 60). Self-report questionnaires designed for young adolescents will be used to assess DE, pubertal status, autonomy difficulties, and mood. Adiposity will be assessed with bioelectrical impedance and body mass index. Hierarchal Linear Models will be used to examine differences in DE by twin type (opposite-or same-sex) and puberty status (pre-pubertal, early puberty, mid-to-late puberty). It is expected that opposite-sex and same-sex twins will not differ significantly on levels of DE during pre-puberty, regardless of twin sex. However, linear relationships between levels of DE and twin type are expected to emerge during and after early puberty, when same-sex male twins are expected to show the most masculinized (i.e., lowest) levels of DE, followed by opposite-sex male twins, opposite-sex female twins, and same-sex female twins. Moreover, these effects will not be accounted for by differences in adiposity, mood, or autonomy difficulties between opposite-sex and same-sex twins. Finally, masculinization of DE in opposite-sex females will not be due to being raised with a brother, as these twins will have lower levels of DE than non-twin females reared with brother. PUBLIC HEALTH RELEVANCE: Findings could necessitate new biological paradigms for understanding the emergence of DE and sex differences in eating disorder risk.